CN112328865A

CN112328865A - Information processing and recommending method, device, equipment and storage medium

Info

Publication number: CN112328865A
Application number: CN201910703743.8A
Authority: CN
Inventors: 占恺峤; 郑东; 周亮
Original assignee: Beijing Dajia Internet Information Technology Co Ltd
Current assignee: Beijing Dajia Internet Information Technology Co Ltd
Priority date: 2019-07-31
Filing date: 2019-07-31
Publication date: 2021-02-05
Anticipated expiration: 2039-07-31
Also published as: CN112328865B

Abstract

The present disclosure relates to an information processing and recommendation method, apparatus, device, and storage medium; the information processing method comprises the following steps: when an information recommendation request sent by a client is received, determining a first social relationship user set of a target account bound by the client and a second social relationship user set of user accounts of the first social relationship user set; determining first interaction parameters of the target account and user accounts of the first social relationship user set and second interaction parameters of the user accounts of the first social relationship user set and user accounts of the second social relationship user set; determining target interaction parameters of the target account and user accounts of the second social relationship user set according to the first interaction parameters and the second interaction parameters; and screening a candidate work set from the published works corresponding to the user account of the second social relation user set according to the target interaction parameters.

Description

Information processing and recommending method, device, equipment and storage medium

Technical Field

The present disclosure relates to computer technologies, and in particular, to an information processing and recommendation method, apparatus, device, and storage medium.

Background

The recommendation system is a service system provided for solving the problem of information overload, can select resources meeting the interest preference of a user from a large amount of information to recommend the resources to the user, and is widely applied to various fields, such as video recommendation, news recommendation, article recommendation, commodity recommendation and the like.

In the related art, a recommendation system generally includes a recall phase and a sort phase. In the recalling stage, the recommendation system performs coarse screening on a large amount of information according to interest preference of a user in certain dimensions so as to screen a smaller candidate set from the large amount of information; in the sorting stage, the recommendation system sorts the information of the candidate set obtained by screening in the recalling stage according to the preference degree of the user to the information and recommends the information to the user.

However, the recall mode adopted by the recommendation system in the recall stage can only be used for screening information depending on the interest preference or the requirement of the user, and the information obtained by screening more reflects the interest dimension of the user. Therefore, the recall method of the related art is single, and information with more dimensions cannot be screened for the user.

Disclosure of Invention

The present disclosure provides an information processing and recommendation method, apparatus, device, and storage medium, to at least solve a problem in the related art that information of more dimensions cannot be obtained for user screening. The technical scheme of the disclosure is as follows:

according to a first aspect of the embodiments of the present disclosure, there is provided an information processing method including:

when an information recommendation request sent by a client is received, determining a first social relationship user set of a target account bound by the client and a second social relationship user set of user accounts of the first social relationship user set;

determining first interaction parameters of the target account and user accounts of the first social relationship user set and second interaction parameters of the user accounts of the first social relationship user set and user accounts of the second social relationship user set; the first interaction parameter and the second interaction parameter are obtained based on social behavior data between the user account and the user account;

determining target interaction parameters of the target account and user accounts of the second social relationship user set according to the first interaction parameters and the second interaction parameters;

and screening a candidate work set from the published works corresponding to the user account of the second social relation user set according to the target interaction parameters.

According to a second aspect of the embodiments of the present disclosure, there is provided an information recommendation method, including:

screening a candidate work set from published works corresponding to the user account of the second social relationship user set according to the target interaction parameters;

ranking the candidate work set to determine a target recalled work set;

and recommending the target recalling work set to the target account.

According to a third aspect of the embodiments of the present disclosure, there is provided an information processing apparatus including:

the system comprises a first determination module, a second determination module and a third determination module, wherein the first determination module is configured to determine a first social relationship user set of a target account bound by a client and a second social relationship user set of user accounts of the first social relationship user set when an information recommendation request sent by the client is received;

a second determination module configured to determine first interaction parameters of the target account and user accounts of the first set of social relationship users, and second interaction parameters of the user accounts of the first set of social relationship users and user accounts of the second set of social relationship users; the first interaction parameter and the second interaction parameter are obtained based on social behavior data between the user account and the user account;

a computing module configured to determine target interaction parameters of the target account and user accounts of the second set of social relationship users according to the first interaction parameters and the second interaction parameters;

and the screening module is configured to screen a candidate work set from published works corresponding to the user account of the second social relationship user set according to the target interaction parameters.

According to a fourth aspect of the embodiments of the present disclosure, there is provided an information recommendation apparatus including:

the screening module is configured to screen a candidate work set from published works corresponding to the user accounts of the second social relationship user set according to the target interaction parameters;

a ranking module configured to rank the set of candidate works to determine a target set of recalled works;

a recommendation module configured to recommend the target recall work set to the target account number.

According to a fifth aspect of embodiments of the present disclosure, there is provided an electronic apparatus including:

a processor;

a memory for storing a computer program executable by the processor;

wherein the processor is configured to carry out the steps of the information processing method and/or the steps of the information recommendation method when executing the computer program.

According to a sixth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, enables the processor to perform the steps of the information processing method and/or the steps of the information recommendation method.

According to a seventh aspect of embodiments of the present disclosure, there is provided a computer program product; the computer program product comprises executable program code which, when loaded into and executed by a processor, enables the processor to perform the steps of the information processing method and/or the steps of the information recommendation method.

The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:

on one hand, for the technical scheme of information processing related to the recall stage, since the user accounts in the second social relationship user set are the user accounts concerned by the target account, generally speaking, the users and the users concerned by the users generally have the same interest, the present disclosure screens out the candidate work set from the published works corresponding to the second degree concerned user accounts of the target account (i.e. the user accounts in the second social relationship user set), so that the candidate work set can not only meet the requirements of the users related to the target account on the interest dimension, but also provide a way for the users related to the target account to know the unknown users who may be interested in, meet the requirements of the users related to the target account on the social dimension, and is favorable for reducing the difficulty of the users related to the target account to know the unknown users who are interested in the interest relationship, and the social circle of the user is enlarged. The candidate work set is screened out based on the interaction parameters of the target account and the second-degree attention user accounts thereof, so that the possibility that the users related to the target account pay attention to the users related to the second-degree attention user accounts which are not established with the social relationship but are close to each other in the social network is improved, the social network of a platform where the users are located is expanded, and the construction of the social relationship among the users is promoted.

On the other hand, for the technical scheme of information recommendation, since the technical scheme of information processing is adopted in the recall stage, the technical scheme of information recommendation has the beneficial technical effects produced by the technical scheme of information processing, and is not described herein again. In addition, the technical scheme of information recommendation screens out the target recall work set for recommending the target account number after sequencing the candidate work sets, so that works which are more in line with the requirements of the users associated with the target account number can be obtained, the social network of the platform where the users are located can be better expanded, and the construction of the social relationship among the users can be promoted.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.

Fig. 1 is a flow chart illustrating an information processing method according to an exemplary embodiment of the present disclosure.

Fig. 2 is a flowchart illustrating an information recommendation method according to an example embodiment of the present disclosure.

Fig. 3 is a block diagram of an information processing apparatus shown in accordance with an exemplary embodiment of the present disclosure.

Fig. 4 is a block diagram illustrating an information recommendation apparatus according to an exemplary embodiment of the present disclosure.

FIG. 5 is a block diagram of an electronic device shown in accordance with an exemplary embodiment of the present disclosure.

Detailed Description

In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

In order to solve at least the problem that information with more dimensions cannot be obtained for user screening in the related art, the embodiment of the disclosure provides an information processing method, a candidate work set is screened out from published works corresponding to a second-degree attention user account of a target account (i.e., a user account in a second social relationship user set), so that the candidate work set can not only meet the requirement of a user associated with the target account on an interest dimension, but also provide a way for the user associated with the target account to know an unknown user who may be interested, meet the requirement of the user associated with the target account on a social dimension, and is beneficial to reducing the difficulty of the user associated with the target account for knowing the unknown user who is associated with the interest, and expanding the social circle of the user. The candidate work set is screened out based on the interaction parameters of the target account and the second-degree attention user accounts thereof, so that the possibility that the users related to the target account pay attention to the users related to the second-degree attention user accounts which are not established with the social relationship but are close to each other in the social network is improved, the social network of a platform where the users are located is expanded, and the construction of the social relationship among the users is promoted.

The information processing method provided by the disclosure can be applied to a recall stage of a recommendation system, and the recommendation system can be applied to a server side with an information recommendation function or information recommendation requirements. The service end can include but is not limited to a service end of at least one user platform of a social platform, a shopping platform, a news platform, a game platform, a reading platform and a learning platform. Each user platform may be a user population oriented client application or web page platform, for example, the social platform may include at least one of, but is not limited to: live broadcast platform, microblog, wechat, QQ.

Next, as shown in fig. 1, fig. 1 is a flowchart illustrating an information processing method according to an exemplary embodiment of the present disclosure, which may be applied to a server, and includes the following steps:

in step S11, when an information recommendation request sent by a client is received, a first social relationship user set of the target account bound by the client and a second social relationship user set of the user accounts of the first social relationship user set are determined.

In step S12, determining first interaction parameters of the target account and the user accounts of the first set of social relationship users, and second interaction parameters of the user accounts of the first set of social relationship users and the user accounts of the second set of social relationship users; the first interaction parameter and the second interaction parameter are obtained based on social behavior data between the user account and the user account.

In step S13, determining target interaction parameters of the target account and the user accounts of the second social relationship user set according to the first interaction parameters and the second interaction parameters.

In step S14, a candidate work set is screened from published works corresponding to the user account of the second social relationship user set according to the target interaction parameter.

In the foregoing, the information recommendation request may carry the target account information, and may be triggered by an operation of viewing information in any platform by a user associated with the target account, for example, in the field of information recommendation, when the user associated with the target account views, through a user platform, a work issued by a user concerned by the user or a work pushed by a server under a discovery column in a live broadcast platform, or views, through an attention column in a microblog, a work issued by the user concerned by the user or a work pushed by the server under a hot column, the generated operation may trigger the relevant platform to send the information recommendation request to the server. After receiving the information recommendation request, the server may send the published works corresponding to all user accounts concerned by the target account to the user platform bound by the target account, and in addition, may also send the published works corresponding to user accounts concerned by all user accounts concerned by the target account to the target account.

In order to recommend the works of the second-degree-of-interest user accounts of the target account to the target account, in the embodiment of the present disclosure, when an information recommendation request sent by a client bound to the target account is received, the target account and all user accounts interested by the target account may be determined according to target account information carried in the information recommendation request, and thus, all user accounts interested by the target account may be stored as a first social relationship user set. After the first social relationship user set is obtained, all user accounts concerned by user accounts of the first social relationship user set may be obtained to obtain a second social relationship user set of the user accounts of the first social relationship user set. The second set of social relationship users may be obtained by a breadth-first search, but the disclosure is not limited thereto.

As such, the number of the first set of social relationship users may be only 1; the number of the second set of social relationship users may be the same as the total number of all user accounts included in the first set of social relationship users, or may be 1.

In addition, in order to clarify the attention relationship between each user account in the first social relationship user set and each user account in the second social relationship user set, an attention mapping relationship between each user account in the first social relationship user set and the user account in the second user relationship set concerned by the user account may be stored.

It should be noted that the first set of social relationship users and the second set of social relationship users may be expressed in a set form or a table form. In addition, for convenience of description, a user account in the first set of social relationship users will be referred to as a first user account, and a user account in the second set of social relationship users will be referred to as a second user account.

After obtaining a first set of social relationship users of the target account and a second set of social relationship users of first user accounts of the first set of social relationship users, for each first user account, a first interaction parameter between the target account and the first user account may be determined according to social behavior data between the target account and the first user account. Similarly, for each second user account concerned by each first user account, a second interaction parameter between the first user account and the second user account may be determined according to social behavior data between the first user account and the second user account. Wherein the social behavior data may include at least one of positive and negative information; the forward information may include at least one of, but is not limited to: click times, praise times, comment times and exchange times; the negative information may include, but is not limited to: the number of times it is annoying. In the following, the meaning of each of the above times is described by taking social behavior data generated by the user account a for the published work corresponding to the user account B as an example:

the number of clicks is used to characterize the total number of times that the user account a views the work of the user account B. The number of praise may be one of the following or a sum or a weighted sum of the following: the number of the works of the user account B contained in the favorite of the user account A and the number of the works of the user account B associated with a mark used for representing the identity or the like of the user account A to the works. The comment times are used for representing the total number of comment information sent by the user account A to the user account B for the works of the user account B. The communication times are used for representing the times of communication establishment between the user account A and the user account B on the current login platform aiming at the works of the user account B or the total number of the chat messages sent to the user account B. The number of offending times can be one of the following or a sum or weighted sum of the following: the total times of canceling the collection operation of the works of the user account B which are collected by the user account A, and the number of the works of the user account B which are associated with marks used for representing that the user account A shows different or annoying marks to the works.

Based on this, in one embodiment, the present disclosure provides a method for determining a first interaction parameter and a second interaction parameter, as follows:

in step S12, for each first user account of the first set of social relationship users, the determining of the first interaction parameter includes:

in step S1211, a corresponding first interaction parameter is calculated according to social behavior data generated by the target account for the published work corresponding to the first user account.

In step S12, for each first user account of the first set of social relationship users, the determining of the second interaction parameter includes:

in step S1212, for each second user account of the second social relationship user set of the first user account, calculating to obtain a corresponding second interaction parameter according to social behavior data generated by the first user account for the published work corresponding to the second user account.

In step S1211, a corresponding first interaction parameter may be calculated according to at least one of positive direction information and negative direction information generated by the target account for the published work corresponding to the first user account. In an example, the first interaction parameter may be calculated based on only the forward information, for example, the value of any item or the sum of the values of any several items in the forward information may be directly used as the first interaction parameter, or the weighted sum of the values of any several items in the forward information may be used as the first interaction parameter; in another example, the first interaction parameter may be calculated based on only the negative information, for example, a value of the negative information, i.e., a value of the number of times of disagreement, may be directly taken as the first interaction parameter. However, any way of calculating the first interaction parameter may cause that the calculated first interaction parameter can only reflect the degree of closeness between the user account and the user account on the front side or the back side, and is relatively lacking in objectivity, so as to solve the technical problem, so as to more comprehensively reflect the degree of closeness between the user account and the user account to obtain a relatively reasonable first interaction parameter, in another example, the first interaction parameter may be calculated by combining the positive information and the negative information, for example, a value of any one of the positive information or a difference between a sum of values of any one of the positive information and a value of the negative information may be used as the first interaction parameter; it is also possible to calculate a weighted sum of values of any several items in the positive information and a product of a value of the negative information and a preset weight, and then use a difference between the weighted sum and the product of the value of the negative information and the preset weight as the first interaction parameter.

Similarly, in step S1212, a corresponding second interaction parameter may be calculated according to at least one of positive direction information and negative direction information, which is generated by the first user account for the published work corresponding to the second user account. In an example, the second interaction parameter may be calculated based on only the forward information, for example, the value of any item or the sum of the values of any several items in the forward information may be directly used as the second interaction parameter, or the weighted sum of the values of any several items in the forward information may also be used as the second interaction parameter; in another example, the second interaction parameter may be calculated based on only the negative information, for example, a value of the negative information, i.e., a value of the number of times of disagreement, may be directly used as the second interaction parameter. However, any way of calculating the second interaction parameter may cause that the calculated second interaction parameter can only reflect the degree of closeness between the user account and the user account on the front side or the back side, and is relatively lacking in objectivity, so as to solve the technical problem, so as to more comprehensively reflect the degree of closeness between the user account and obtain a relatively reasonable second interaction parameter, in another example, the second interaction parameter may be calculated by combining the positive information and the negative information, for example, a value of any item in the positive information or a difference between a sum of values of any item and a value of the negative information may be used as the second interaction parameter; it is also possible to calculate a weighted sum of values of any several items in the positive information and a product of a value of the negative information and a preset weight, and then use a difference between the weighted sum and the product of the value of the negative information and the preset weight as the second interaction parameter.

However, in order to more objectively and accurately represent the social situation between the user and the user to better improve the rationality of the screened candidate works set, in another embodiment, the present disclosure provides another technical solution for calculating corresponding interaction parameters based on social behavior data, as follows:

in the step S1211, calculating a corresponding first interaction parameter according to social behavior data generated by the target account for the published work corresponding to the first user account may include the following steps:

in step S12111, respectively calculating a click rate, an approval rate, a comment rate, an offense rate, and a communication rate, which correspond to the published work generated by the target account for the first user account;

in step S12112, a corresponding first interaction parameter is calculated according to at least one of the click rate, the like rate, the comment rate, the traffic rate, and the annoying rate calculated in step S12111.

Similarly, in the step S1212, for each second user account of the second social relationship user set of the first user account, calculating to obtain a corresponding second interaction parameter according to social behavior data generated by the first user account for the published work corresponding to the second user account, which may include the following steps:

in step S12121, respectively calculating a click rate, an approval rate, a comment rate, an offense rate, and a communication rate corresponding to the published work corresponding to the second user account based on the number of clicks, the number of likes, the number of comments, the number of offences, and the number of communications generated by the first user account;

in step S12122, a corresponding second interaction parameter is calculated according to at least one of the click rate, the like rate, the comment rate, the traffic rate, and the annoying rate calculated in step S12121.

In step S12111, regarding a click rate, the click rate may be equal to a ratio of the number of clicks to the number of times that the published work corresponding to the first user account is sent to the target account, for example, if there are N published works corresponding to the first user account and each work is sent to the target account 2 times, it may be considered that the number of times that the published work corresponding to the first user account is sent to the target account is N × 2 — 2N times; based on this, if the target account clicks and views 1 time on each released work corresponding to the first user account, it may be considered that the corresponding number of clicks is N × 1 ═ N times; as a result, the click rate is N/2N 0.5. For the like rate, the like rate may be equal to a ratio of the number of like times to the number of click times, and based on this, the like rate may be further understood in combination with the above explanation example of the click rate. For the comment rate, the comment rate may be equal to a ratio of the number of comments to the number of clicks, and based on this, the comment rate may be further understood in combination with the above explanation example of the click rate. For the exchange rate, the exchange rate may be equal to a ratio of the exchange number to the click number, based on which the exchange rate may be further understood in connection with the above explanation example of the click rate. For the disagreement rate, the disagreement rate may be equal to a ratio of the disagreement number to the click number, based on which the disagreement rate may be further understood in connection with the above explanation example of the click rate.

In addition, the meaning and the calculation principle of the click rate, the like rate, the comment rate, the traffic rate and the annoying rate in step S12121 can be known based on the above description, and will not be described herein.

It should be noted that the statistical time period of any item of data in the social behavior data may be a time period from when the user account and the user account successfully establish a social relationship to the current time, or may be a preset time period from when the user account and the user account successfully establish a social relationship to the current time, where the preset time period may be set according to actual needs or experience, and this is not limited by the present disclosure.

Based on the previous embodiment, in an embodiment, for each first user account of the first social relationship user set, the first interaction parameter is equal to a first weighted sum of a corresponding click rate, a corresponding approval rate, a corresponding comment rate and a corresponding traffic rate, or equal to a difference between the first weighted sum and a product of weights corresponding to the approval rate and the disapproval rate, so as to simplify a calculation process of the first interaction parameter and the second interaction parameter, reduce a calculation difficulty, and further improve an information recall efficiency, and improve accuracy of the first interaction parameter and the second interaction parameter. Similarly, for each first user account of the first set of social relationship users and each second user account of the second set of social relationship users of the first user accounts, the second interaction parameter is equal to a second weighted sum of the corresponding click rate, like rate, comment rate, and traffic rate, or equal to a difference of the second weighted sum and a product of weights corresponding to the like rates. In order to further simplify the calculation process, in one example, the weighting coefficients of the traffic rate and the nuisance rate may be set to 0, which is equivalent to calculating the interaction parameters by using only the click rate, the approval rate and the comment rate.

It should be noted that the first interaction parameter and the second interaction parameter are both used for representing the degree of closeness between the corresponding user and the user, and are in positive correlation with the degree of closeness.

Therefore, corresponding weight coefficients can be distributed to the click rate, the like rate, the comment rate, the exchange rate and the disagreement rate according to actual conditions, so that the occupation ratios of all numerical values in the intimacy degree are balanced, the rationality and the accuracy of the first interaction parameter and the second interaction parameter obtained through calculation are improved, and the calculation processes related to the calculation mode of weighted sum/difference are fewer, so that the calculation process of the first interaction parameter and the second interaction parameter can be simplified, the calculation difficulty is reduced, and the information recall efficiency is improved.

Although the first interaction parameter and the second interaction parameter with higher rationality and higher accuracy can be calculated through the description in the previous embodiment, because the user accounts and the concerned user accounts have different degrees of closeness, the calculated first interaction parameter or second interaction parameter may be too large or too small, and thus may have adverse effects on the calculation of the target interaction parameter, such as increasing the calculation complexity of the target interaction parameter, and therefore, in another embodiment, the present disclosure provides another technical scheme for calculating the corresponding interaction parameter based on social behavior data, as follows:

in step S12111', a click rate, an approval rate, a comment rate, an offense rate, and a communication rate are respectively calculated based on the number of clicks, the number of likes, the number of comments, the number of offense times, and the number of exchanges generated by the target account for the published work corresponding to the first user account;

in step S12112 ', a first weighted sum of the click rate, the like rate, the comment rate, and the traffic rate obtained by the step S12111' is calculated, and a first difference of the first weighted sum and a product of weights corresponding to the disagreement rate and the disagreement rate is calculated;

in step S12113', the first difference is converted into a value in the interval [0, 1] to obtain a corresponding first interaction parameter.

Similarly, in the step S1212, for each second user account of the second social relationship user set of the first user account, calculating to obtain a corresponding second interaction parameter according to social behavior data generated by the first user account for the published work corresponding to the second user account, which may include the following steps: in step S12121', the click rate, the like rate, the comment rate, the disagreement rate, and the exchange rate are respectively calculated based on the number of clicks, the like number, the comment number, the disagreement number, and the exchange number of times that the first user account generates with respect to the published work corresponding to the second user account;

in step S12122 ', a first weighted sum of the click rate, the like rate, the comment rate, and the traffic rate obtained by the step S12121' is calculated, and a second difference of the first weighted sum and a product of weights corresponding to the disagreement rate and the disagreement rate is calculated;

in step S12123', the second difference is converted into a value in the interval [0, 1] to obtain a corresponding second interaction parameter.

In an example, converting the first difference value or the second difference value into a value in the interval [0, 1] can be implemented by the following formula (i):

min (max (a × ctr + b × ltr + c × cmtr + d × Int-e × dil, 0), 1) -formula (i)

In the formula I, M represents an interaction parameter; ctr represents the click rate, and a represents the weight coefficient of the click rate; ltr represents the like rate, b represents the weight coefficient of the like rate; cmtr represents a comment rate, and c represents a weight coefficient of the comment rate; int denotes the traffic rate, d denotes the weighting factor of the traffic rate; dil represents the frequency of the offensiveness, and e represents the weight coefficient of the offensiveness.

In another embodiment, if the first interaction parameter and the second interaction parameter do not need to be calculated according to the negative information, the weighting factor of the annoying rate can be set to zero, based on which the step S12112' can be adaptively adjusted to: calculating a first weighted sum of the click rate, the like rate, the comment rate, and the traffic rate obtained through the step S12111'; step S12113' is adjusted accordingly to: the first weighted sum is converted into a value in the interval [0, 1] to obtain a corresponding first interaction parameter. Similarly, the step S12122' may be adaptively adjusted to: calculating a first weighted sum of the click rate, the like rate, the comment rate, and the traffic rate obtained through the step S12121'; step S12123' is adjusted accordingly to: the second weighted sum is converted into a value in the interval [0, 1] to obtain a corresponding second interaction parameter. Accordingly, the formula (r) can be adaptively adjusted to: m ═ min (max (a × ctr + b × ltr + c × cmtr + d × Int, 0), 1).

Therefore, the value of the first interaction parameter and the value of the second interaction parameter can be in the interval [0, 1] through the min function and the max function.

After the first interaction parameter and the second interaction parameter are calculated and obtained through any one of the calculation modes of the first interaction parameter and the second interaction parameter, the target interaction parameter between the target account and each second user account of the second social relationship user set can be calculated and obtained based on the first interaction parameter and the second interaction parameter. In an embodiment, for each second user account, the calculating of the target interaction parameter may include: for each first user account with a social relationship with the second user account, calculating the product of a first interaction parameter and a second interaction parameter corresponding to the first user account; and calculating the sum of all the obtained products to obtain the target interaction parameter.

The following describes, for example, a calculation process of target interaction parameters between the target account and any second user account:

assuming that the total number of first user accounts in the first social relationship user set is n, the total number of second user accounts in the second social relationship user set is m, and a first user account of a social relationship exists between the first social relationship user set and an ith second user account in the second social relationship user setThe total number of the numbers is i (it can also be understood that the total number of the first user accounts concerning the jth second user account in the second social relationship user set is i), wherein i is an integer and is greater than or equal to 1 and less than or equal to n, and j is an integer and is greater than or equal to 1 and less than or equal to m. Based on this, in the i first user accounts, the first interaction parameter of the kth first user account and the target account can be represented as M_k，1The second interaction parameter of the kth first user account and the jth second user account can be expressed as

k is an integer and is more than or equal to 1 and less than or equal to i. Based on this, for the jth second user account, the target interaction parameter D between the jth second user account and the target account can be calculated by the following formula-_j：

Therefore, the target interaction parameters between the target account and each second user account can be obtained through calculation according to a formula II.

After target interaction parameters of all second user accounts in a second social relationship user set and the target account are obtained through calculation, a candidate user set can be determined from the second social relationship user set according to all the calculated target interaction parameters, and published works corresponding to the user accounts in the candidate user set are used as a candidate work set. In an example, a published work corresponding to a second user account corresponding to a target interaction parameter larger than a first preset threshold may be used as a candidate work set, where the first preset threshold may be preset according to actual needs or experience or experiments. In another example, all the target interaction parameters may be sequentially ranked in a descending order, and the published works corresponding to the second user accounts corresponding to the target interaction parameters ranked in the top preset number are used as the candidate work set. In another example, all the target interaction parameters may be sequentially ranked from small to large, and the published works corresponding to the second user accounts corresponding to the ranked preset number of target interaction parameters are used as the candidate work set.

Although any embodiment of screening candidate works according to the target interaction parameters can improve the possibility that the target account pays attention to a second user account which is not established with the attention relationship but is close to the second user account in the social network, the social network of the platform where the user is located is expanded, and the establishment of the social relationship among the users is promoted. However, because the change of the degree of affinity between the target account and the second user account in a short period is small, the method of screening and obtaining the candidate work set only according to the size of the target interaction parameter may cause that works which are already published by the same batch of second user accounts are recommended to the target account in each recommendation in a short period, so that the second user account works with a lower degree of affinity do not have a chance of being revealed, and the construction of the social network in the platform is limited to a certain extent. Therefore, in order to solve the technical problem, while ensuring that the probability of the selected work corresponding to the second user account is in a positive correlation with the target interaction parameter corresponding to the second user account, the work corresponding to the second user account with a smaller value also has a chance of being revealed, and the limitation of social network construction and information recommendation is broken through, in an embodiment, the present disclosure provides another implementation manner of screening a candidate work set according to the target interaction parameter, in this embodiment, in step S14, a reservoir sampling method is added, which can be understood as: the screening of the candidate work set from the published works corresponding to the user account of the second social relationship user set according to the target interaction parameter comprises the following steps:

in step S141, for each user account of the second set of social relationship users, a random number corresponding to the user account is generated.

In step S142, for each user account of the second social relationship user set, calculating a candidate reference value for representing the size of the probability that the user is selected according to the random number corresponding to the user account and the target interaction parameter; the candidate reference value and the target interaction parameter are in positive correlation;

in step S143, a candidate user set is determined from the second social relationship user set according to the calculated candidate reference value, and the published works corresponding to the user accounts of the candidate user set are used as a candidate work set.

Therefore, by applying a reservoir sampling method, a candidate reference value used for representing the size of the probability of the second user account being selected is calculated based on the random number generated for the second user account and the target interaction parameter corresponding to the second user account, and the candidate reference value and the target interaction parameter are limited to be in positive correlation in the calculation process, so that the probability of the second user account being selected and the target interaction parameter corresponding to the second user account having a smaller value can be ensured to be in positive correlation by combining the random number and the target interaction parameter, and meanwhile, the works of the second user account corresponding to the target interaction parameter having a smaller value also have a chance of being revealed.

In the above, the generation of the random number may refer to related technologies, and will not be described herein.

In the present disclosure, to simplify the calculation steps, one random number is generated for each second user account in the interval (0, 1). Based on this, the candidate reference value of each second user account can be calculated by the following formula (c):

in the formula III, K_jA candidate reference value, T, representing a jth second user account in the second set of social relationship users_jA random number, D, representing the corresponding account number of the jth second user in the second social relationship user set_jAnd representing a target interaction parameter corresponding to the jth second user account in the second social relationship user set. Therefore, the candidate reference value of each second user account can be calculated through the formula (c).

After calculating the candidate reference value of each second user account, in an example, in step S143, the determining process of the candidate user set may include: and determining users corresponding to the candidate reference value which is greater than or equal to a second preset threshold value from the second social relationship user set to obtain a candidate user set, wherein the second preset threshold value can be preset according to actual requirements or experience or experiments. In another example, in step S143, the determining of the candidate user set may include: sequencing all the calculated candidate reference values in sequence according to the sequence from big to small; and selecting the user accounts corresponding to the top preset number of candidate reference values from the second social relationship user set to obtain a candidate user set. In yet another example, in step S143, the determining of the candidate user set may include: sequencing all the calculated candidate reference values in sequence according to the sequence from small to large; and selecting the user accounts corresponding to the sorted preset number of candidate reference values from the second social relation user set to obtain a candidate user set.

After the candidate user set is obtained by any of the above embodiments, the published works corresponding to the user accounts in the candidate user set may be used as the candidate work set.

It should be noted that the embodiment of the present disclosure is not limited to determining the candidate work set by the reservoir sampling method, for example, if it is not necessary to ensure that the probability that the second user account is selected has a positive correlation with the target interaction parameter, but other targets need to be achieved, then other sampling methods may be used to achieve the determination of the candidate work set.

In addition, because social relationships that concern each other may exist between the user accounts, there may be a possibility that a target account is included in a second set of social relationship users of the user accounts of the first set of social relationship users, and then, a situation that a work corresponding to the target account is recommended back to the target account may occur, which not only affects user experience, but also increases system workload. Therefore, to solve this technical problem, in an embodiment, before the step S12, the method may further include: determining whether the target account exists in the second set of social relationship users; and if the target account exists in the second social relationship user set, deleting the target account from the second social relationship user set. Therefore, the second set of social relationship users in step S12 may not include the target account, so as to avoid recommending the work of the target account back to the target account.

In addition, in one aspect, in the second social relationship user set, since there may be some cases where the fans of the second user account are many, such as some large-V users, it may be understood that there is a scholar, celebrity, red or star, etc. that has a certain influence on the social network. The large V users can easily acquire the interaction behaviors of fans compared with other common users, and can be easily concerned by users related to user accounts in a first social relationship user set concerned by the target account, so that the target interaction parameters corresponding to the accounts of the large V users are higher, a large amount of traffic is easily concentrated to second social relationship users with a large number of fans, and the construction of a social network is uneven. On the other hand, in the second social relationship user set, there may be second user accounts with a large number of fans or a large number of reported times, and if works corresponding to the second user accounts which are reported many times are recommended to the target account, the health construction of the social network may be influenced. Therefore, in order to solve at least one of the technical problems, in an embodiment, before the step S14, the target interaction parameter is modified, and it is understood that, before the candidate work set is screened from the published works corresponding to the user accounts of the second social relationship user set according to the target interaction parameter, the method may further include: and for each target interaction parameter, updating the target interaction parameter according to at least one of the total number of user accounts with social relationships in the second user account of the second social relationship user set corresponding to the target interaction parameter and the total number of reports of the second user account by other user accounts. The total number of user accounts with social relationships to the second user account of the second social relationship user set corresponding to the target interaction parameter may be understood as the total number of fan of the second user account, and the total number of reported second user accounts by other user accounts may be understood as the total number of reported second user accounts, or may be understood as the total number of notification messages that the server receives to indicate that the second user account is reported. Thus, the target interaction parameter used in step S14 may be the updated target interaction parameter.

Under the requirement of avoiding concentration of a large amount of traffic to the second social relationship users with a large number of fans, for each target interaction parameter, the target interaction parameter can be updated only based on the total number of fans of the second user account corresponding to the target interaction parameter, for example, the target interaction parameter can be updated through a formula (r) -a formula (r-)

Updating the target interaction parameter, wherein D'_j1Representing target interaction parameters after j second user account updating in the second social relationship user set; d_jRepresenting target interaction parameters before updating of the jth second user account; f_jRepresenting the total number of fans of the jth second user account; f represents a constant, which is used as a smoothing term of the denominator and can be preset according to actual needs or experience or experiments, and is not described herein.

Under the requirement that only the second user account reported for multiple times is prevented from causing adverse effects on the health construction of the social network, for each target interaction parameter, the target interaction parameter can be updated only based on the reported total number of the second user accounts corresponding to the target interaction parameter, for example, the target interaction parameter can be updated through a formula (v) -c

Updating the target interaction parameter, wherein D'_j2Representing a jth second user in the second set of social relationship usersUpdating the target interaction parameters by the account; d_jRepresenting target interaction parameters before updating of the jth second user account; g_jRepresenting the reported total number of the jth second user account; g represents a constant, which is used as a smoothing term of the denominator, and can be preset according to actual needs or experience or experiments, which is not described herein.

Under the requirement of avoiding adverse effects on the health construction of the social network caused by the second social relationship user set with a large amount of traffic and a large number of fans and the second user account reported for many times, for each target interaction parameter, the target interaction parameter can be updated only based on the total number of fans of the second user account corresponding to the target interaction parameter and the total number reported, for example, through a formula | -

Updating the target interaction parameter, wherein D'_jRepresenting target interaction parameters after j second user account updating in the second social relationship user set; d_jRepresenting target interaction parameters before updating of the jth second user account; f_jRepresenting the total number of fans of the jth second user account; g_jRepresenting the reported total number of the jth second user account; o represents a constant, which is used as a smoothing term of the denominator, and can be preset according to actual needs or experience or experiments, which is not described herein.

It should be noted that the update of the target interaction parameter in the present disclosure is not limited to any one of the above update manners, for example, in other embodiments, only the target interaction parameter corresponding to the second user account whose total number of fans exceeds the first threshold or whose total number of reported fans exceeds the second threshold may be updated; or, the target interaction parameter can be updated through other formulas except for formulas (iv) to (sixty), so long as the updated target interaction parameter is in a reasonable numerical range relative to the target interaction parameter before updating, and the purpose of correction can be achieved.

However, before the target interaction parameters are updated, if the consumption experience is optimized to the utmost extent and the demand that the flow is concentrated on a small part of users with high-quality second social relations is not respected, it is considered that recommending the published works corresponding to the second user account with the high fan count to the target account is reasonable, and at this time, the influence of the fan count of the second user account can be ignored, so that the target interaction parameters are not updated, or the target intimacy is updated only according to the reported total number of the second user account. On the contrary, if the influence of the second user accounts with high total number of fans of the head producer, such as fans, is weakened and the demand of building an equal community atmosphere is biased, the formula (iv) or (iv) can be adjusted to increase the pressure on the second user accounts with excessive total number of fans, for example, the denominator in the formula (iv) or (iv) is increased, so that the updated target interaction parameter is reduced relative to the target interaction parameter before updating. Of course, the update strategy of the target interaction parameter may also be adjusted according to other requirements.

On the other hand, in order to sort the candidate work sets and then screen out the target recall work set for recommending the target account number, so as to obtain works which better meet the requirements of the target account number, better expand the social network of the platform where the user is located and promote the construction of the social relationship among the users, based on the information processing method provided by any one of the embodiments, the disclosure also provides an information recommending method including the information processing method provided by any one of the embodiments.

As shown in fig. 2, fig. 2 is a flowchart illustrating an information recommendation method according to an exemplary embodiment, where the information recommendation method is applicable to a server, and includes the following steps:

in step S21, when an information recommendation request sent by a client is received, a first social relationship user set of the target account bound by the client and a second social relationship user set of the user accounts of the first social relationship user set are determined.

In step S22, determining first interaction parameters of the target account and the user accounts of the first set of social relationship users, and second interaction parameters of the user accounts of the first set of social relationship users and the user accounts of the second set of social relationship users; the first interaction parameter and the second interaction parameter are obtained based on social behavior data between the user account and the user account.

In step S23, target interaction parameters of the target account and the user accounts of the second social relationship user set are determined according to the first interaction parameters and the second interaction parameters.

In step S24, a candidate work set is screened from published works corresponding to the user account of the second social relationship user set according to the target interaction parameter.

In step S25, the candidate work set is subjected to ranking processing to determine a target recall work set.

In step S26, the target recall work set is recommended to the target account number.

For the description of the steps S21 to S24, refer to the technical solution described in the information processing method, which is not repeated herein. The understanding of the steps S25 and S26 can refer to the processing principle of the recommendation system in the related art in the sorting stage, and will not be described herein.

In an embodiment, before the step S26, a step of reordering the target recall work set may be added to increase the diversity of the works finally recommended to the target account.

It should be noted that the information recommendation method provided in the embodiment of the present disclosure is applied to the information processing method provided in any one of the above embodiments of the present disclosure at a recall stage, so that the information recommendation method has the beneficial technical effects produced by the information processing method described in any one of the above embodiments. And the user information and the account information related to the scheme are acquired by authorization of the user and are subjected to subsequent processing and analysis.

Corresponding to the aforementioned information processing method, the present disclosure also provides an information processing apparatus, as shown in fig. 3, fig. 3 is a block diagram of an information processing apparatus shown according to an exemplary embodiment of the present disclosure, which includes a first determining module 31, a second determining module 32, a calculating module 33, and a filtering module 34.

The first determining module 31 is configured to determine, when an information recommendation request sent by a client is received, a first set of social relationship users of a target account bound by the client and a second set of social relationship users of user accounts of the first set of social relationship users.

The second determining module 32 is configured to determine first interaction parameters of the target account and user accounts of the first set of social relationship users, and second interaction parameters of the user accounts of the first set of social relationship users and user accounts of the second set of social relationship users; the first interaction parameter and the second interaction parameter are obtained based on social behavior data between the user account and the user account.

The calculation module 33 is configured to determine target interaction parameters of the target account and user accounts of the second set of social relationship users according to the first interaction parameters and the second interaction parameters.

The screening module 34 is configured to screen a candidate work set from published works corresponding to the user accounts of the second social relationship user set according to the target interaction parameter.

In an embodiment, the second determination module 32 includes a first calculation unit and a second calculation unit.

The first calculation unit is configured to calculate, for each user account of the first social relationship user set, a corresponding first interaction parameter according to social behavior data generated by the target account for a published work corresponding to the user account.

The second calculation unit is configured to calculate, for each user account of the first social relationship user set and for each second user account of the second social relationship user set of the user accounts, a corresponding second interaction parameter according to social behavior data generated by the user account for a published work corresponding to the second user account.

In one embodiment, the information processing apparatus further includes an update module.

The updating module is configured to update the target interaction parameters according to at least one of the total number of user accounts having social relationships with the second user account of the second social relationship user set corresponding to the target interaction parameters and the total number of the second user accounts reported by other user accounts before the candidate work set is screened by the screening module 34, so that the screening module 34 screens the candidate work set from the published works corresponding to the user accounts of the second social relationship user set according to the updated target interaction parameters.

In an embodiment, the screening module 34 comprises a generating unit, a third calculating unit and a determining unit.

The generation unit is configured to generate a random number corresponding to each user account of the second social relationship user set.

The third calculation unit is configured to calculate, for each user account of the second social relationship user set, a candidate reference value used for representing the size of the probability that the user account is selected according to the random number corresponding to the user account and the target interaction parameter; the candidate reference value and the target interaction parameter have positive correlation.

The determining unit is configured to determine a candidate user set from the second social relationship user set according to the calculated candidate reference value, and take a published work corresponding to a user account of the candidate user set as a candidate work set.

In one embodiment, the determining unit includes a sorting subunit and a selecting subunit.

The sorting subunit is configured to sort all the calculated candidate reference values sequentially in descending order.

The selecting subunit is configured to select, from the second social relationship user set, user accounts corresponding to a preset number of candidate reference values ranked in the top order, so as to obtain a candidate user set.

Corresponding to the foregoing information recommendation method, the present disclosure further provides an information recommendation apparatus including the information processing apparatus provided in any of the foregoing embodiments, as shown in fig. 4, fig. 4 is a block diagram of an information recommendation apparatus shown in the present disclosure according to an exemplary embodiment, where the information recommendation apparatus includes a first determining module 41, a second determining module 42, a calculating module 43, a screening module 44, a sorting module 45, and a recommending module 46.

The first determining module 41 is configured to determine, when an information recommendation request sent by a client is received, a first set of social relationship users of a target account bound by the client and a second set of social relationship users of user accounts of the first set of social relationship users.

The second determining module 42 is configured to determine first interaction parameters of the target account and the user accounts of the first set of social relationship users, and second interaction parameters of the user accounts of the first set of social relationship users and the user accounts of the second set of social relationship users; the first interaction parameter and the second interaction parameter are obtained based on social behavior data between the user account and the user account.

The calculation module 43 is configured to determine target interaction parameters of the target account and user accounts of the second set of social relationship users according to the first interaction parameters and the second interaction parameters.

The screening module 44 is configured to screen a candidate work set from published works corresponding to the user accounts of the second social relationship user set according to the target interaction parameter.

The ranking module 45 is configured to rank the set of candidate works to determine a target set of recalled works;

the recommendation module 46 is configured to recommend the target set of recalled works to the target account number.

It should be noted that the information recommendation apparatus has the same functions as the information processing apparatus except for the sorting module 45 and the recommendation module 46. In addition, since the information recommendation device includes the information processing device in any of the embodiments, the information recommendation device has all modules, all units, and all sub-units included in the information processing device in any of the embodiments, which are not described herein again.

With regard to the apparatuses in the above embodiments, the specific manner in which the respective modules and the respective units perform operations has been described in detail in the embodiments related to the method, and will not be described in detail here. Also, the above-described embodiments of the apparatus are merely illustrative, wherein the modules and/or units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units.

Corresponding to the embodiment of the method, the electronic equipment is further provided by the disclosure. In one example, the electronic device may be provided as a server, as shown in fig. 5, fig. 5 being a block diagram of an electronic device shown in accordance with an exemplary embodiment of the present disclosure. The electronic device 500 comprises a processing component 522 and a memory 532.

Wherein the memory 532 is used for storing computer programs executable by the processing component 522; the processing component 522 is configured to implement the information processing method in any of the above embodiments and/or the information recommendation method in any of the above embodiments when executing the computer program.

In an embodiment, the processing component 522 may include one or more processors.

In one embodiment, the memory 532 may store memory resources other than the computer programs described above, such as application programs. The application programs stored in the memory 532 may include one or more modules corresponding to a set of instructions.

In an embodiment, the electronic device 500 may further include a power component 526. The power component 526 may be configured to perform power management operations of the electronic device 500.

In an embodiment, the electronic device 500 may also include a wired or wireless network interface 550, and an input-output (I/O) interface 558. The network interface 550 is configured to connect the electronic device 500 to a network.

In addition, the electronic device 500 may operate an operating system stored in the memory 532, such as Android, IOS, Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.

Corresponding to the aforementioned method embodiments, in the exemplary embodiment, there is also provided a computer-readable storage medium comprising a computer program, such as the memory 532 comprising the computer program. The computer program may be executed by the processing component 522 of the electronic device 500 to perform the information processing method in any of the embodiments and/or the information recommendation method in any of the embodiments.

The computer-readable storage medium may include: permanent or non-permanent removable or non-removable media. The information storage functionality of the computer-readable storage medium may be implemented by any method or technology that may be implemented. The information may be computer readable instructions, data structures, models of programs, or other data.

Additionally, the computer-readable storage media include, but are not limited to: phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology memory, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or other non-transmission media that can be used to store information that can be accessed by a computing device.

In an exemplary embodiment, the present disclosure further provides a computer program product for executing the steps of the information processing method in any of the above embodiments and/or the information recommendation method in any of the above embodiments. The computer program product includes executable program code. After the processing component 522 of the electronic device 500 loads and executes the program code, the information processing method in any of the above embodiments and/or the information recommendation method in any of the above embodiments may be executed to implement corresponding functions.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An information processing method characterized by comprising:

2. The method of claim 1, wherein the determining of the first interaction parameter for each user account of the first set of social relationship users comprises:

and calculating to obtain a corresponding first interaction parameter according to social behavior data generated by the target account aiming at the published works corresponding to the user account.

3. The method of claim 1, wherein the determining of the second interaction parameter for each user account of the first set of social relationship users comprises:

and for each second user account of the second social relationship user set of the user accounts, calculating to obtain a corresponding second interaction parameter according to social behavior data generated by the user account aiming at the published works corresponding to the second user account.

4. The method of claim 1, wherein prior to the screening of a candidate set of works from published works corresponding to user accounts of the second set of social relationship users according to the target interaction parameter, the method further comprises:

and for each target interaction parameter, updating the target interaction parameter according to at least one of the total number of user accounts with social relationships in the second user account of the second social relationship user set corresponding to the target interaction parameter and the total number of reports of the second user account by other user accounts.

5. The method according to any one of claims 1 to 4, wherein the step of screening out a candidate work set from published works corresponding to user accounts of the second social relationship user set according to the target interaction parameter comprises:

generating a random number corresponding to each user account of the second social relationship user set;

for each user account of the second social relationship user set, calculating to obtain a candidate reference value for representing the selected probability of the user account according to the random number corresponding to the user account and the target interaction parameter; the candidate reference value and the target interaction parameter are in positive correlation;

and determining a candidate user set from the second social relationship user set according to the calculated candidate reference value, and taking the published works corresponding to the user accounts of the candidate user set as the candidate work set.

6. An information recommendation method, comprising:

ranking the candidate work set to determine a target recalled work set;

and recommending the target recalling work set to the target account.

7. An information processing apparatus characterized by comprising:

8. An information recommendation apparatus, comprising:

9. An electronic device, comprising:

a processor;

a memory for storing a computer program executable by the processor;

wherein the processor is configured to, upon execution of the computer program, implement the steps of the information processing method of any one of claims 1 to 5 and/or the steps of the information recommendation method of claim 6.

10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, enables the processor to carry out the steps of the information processing method according to one of claims 1 to 5 and/or the steps of the information recommendation method according to claim 6.